Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    say Capitalized names

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     holders
    -0.07
     formed
    -0.07
     stress
    -0.06
     simulations
    -0.06
     tears
    -0.06
    thew
    -0.06
    $view
    -0.06
    Site
    -0.06
    $html
    -0.06
     //----------------
    -0.06
    POSITIVE LOGITS
     BorderSide
    0.06
    れない
    0.06
    ЕН
    0.06
    γο
    0.06
     showed
    0.06
    リス
    0.06
     支
    0.06
     bert
    0.06
     Mog
    0.06
    _PLAYER
    0.06
    Act Density 0.003%

    No Known Activations